Symbiotic associations between
two anthozoans and crustaceans
in the Koster fiord area
Lisbeth Jonsson
Master Thesis in Marine Zoology, 20 points
Department of Marine Ecology, Göteborg University
Supervisor: Tomas Lundälv
Contribution nr 67
Assistant supervisor: Prof. Kerstin Johannesson
Examiner: Prof. Rutger Rosenberg
Date: 21 october 1998
Abstract:
While symbiotic associations between sea anemones and
crustaceans appear to be common in tropical waters, few such associations are
known from temperate waters, except for the symbiosis between hermit crabs
and sea anemones. However, for the last two years observations with ROVs
(remotely operated vehicles) in the Koster fiord area have suggested that
shrimps of certain species (Spirontocaris liljeborgii, Lebbeus polaris, Pandalus
borealis, P. propinquus, P. montagui) associate with the sea anemone
Bolocera tuediae and the cerianthid Pachycerianthus multiplicatus by
aggregating beneath their tentacles. The lithodid crab Lithodes maja was also
found to associate with B. tuediae. In this study, the hypotheses regarding
anthozoan-crustacean associations were tested using sequences of video
recording by ROVs in situ and laboratory experiments. The results from both
the field study and the experiments showed that the crustaceans did have an
association with the two anthozoans. The association is presumably a
facultative commensalistic association as the crustaceans are all found living
as non-symbionts on the sea bottom as well, and the two anthozoans have
not to date been perceived as gaining any advantages of the association. S.
liljeborgii has the closest association with both anthozoans, while B. tuediae
is the preferred host of all associating species. Reasons for the associations are
assumed to be protection from predators, and in the case of the shrimps also
an easy access to a food source.
Introduction
In tropical waters symbiotic associations between sea anemones, and fishes
and crustaceans are frequent.
In temperate waters in the north-eastern Atlantic and the North Sea the
symbiosis between hermit crabs and the sea anemones carried on their shells
is well-known, however reports of other associations between sea anemones
and crustaceans in these waters have been very few indeed. Vader (1970a)
working in Norwegian waters, found amphipods living in the gastrovascular
cavity of the sea anemone Bolocera tuediae, and copepods associated with the
same sea anemone but living as parasites in large, roundish galls, formed
from the mesentery walls of the host (Vader 1970b).
The small numbers of associations found between sea anemones and
crustaceans in temperate waters might be due to different conditions in these
waters which could make symbiosis a less adaptive strategy than in tropical
waters. Fautin et al (1995) proposed that the associations between tropical sea
anemones and shrimps are a result of adaptations to the particular conditions
found in the tropical waters. Another explanation might simply be that
observations have been more difficult to make in temperate waters so that,
among other things, many symbiotic associations yet have to be discovered.
1
Shrimp-actinian symbiosis are usually viewed as commensalistic, the
shrimps deriving benefits (food scraps and protection) from their hosts but
having no obvious effect on them (Shick 1991).
A few associations have been found to be mutualistic, e.g. in tropical waters
between the snapping shrimp Alpheus armatus and the sea anemone
Bartholomea annulata, where the shrimp chases away the polychaete
Harmodice carunculata which otherwise would predate upon the sea
anemone (Smith 1977). The shrimps are typically found near the column of
the sea anemone, protected by the tentacles where they are seen feeding on
detritus. Both the growth rate and size of the shrimps were found to correlate
with the size of the host sea anemone (Knowlton 1979).
Other associations have been found to be parasitic, e.g. between the shrimp
Periclimenes brevicarpalis and its host sea anemones (at least eight species
have been observed as hosts). Both in field and laboratory studies the shrimp
has been seen to tear tentacles from its host for food, when deprived of other
food sources. In laboratory conditions, the sea anemones would frequently die
from this tentacle loss while in the field no “heavy damage” was observed in
the host sea anemones. An association that commonly would result in a sea
anemone with extremely clipped tentacles, would have to be unstable, so
presumably the shrimps “knew” how many tentacles could be tore off before
fatally damaging the host sea anemone during normal conditions. The
growth, longevity and fecundity of P. brevicarpalis, even in the absence of
predators, was enhanced by living with its host sea anemone (Fautin et al
1995).
In the tropics most shrimp species associated with sea anemones are found
with only some of the species sympatric with them. This implies that shrimps
are host species-specific (i.e. able to distinguish between potential hosts and
non hosts) (Bruce 1976b in Guo et al 1996).
Guo et al (1996) performed laboratory experiments where shrimps of three
species, Periclimenes ornatus, P. brevicarpalis and Thor amboinensis had to
locate their host sea anemone in a Y maze. The result showed that none of
the shrimps located a sea anemone by visual cues alone. Instead a significant
proportion of the shrimps of all three species were chemically attracted to
their host sea anemones, but not to any other species of sea anemones.
Shrimps seem to detect a sea anemone from a distance by diffusing
compounds, then switch to cues such as vision, touch, and contact
chemosensation (gustation) near the sea anemone.
Many of the tropical symbiotic shrimps and fishes have an obligate
association with their sea anemones and are consequently never found
without a host (Crawford 1992).
Lysmata grabhami, a shrimp found around Madeira and Canary Islands, has
been observed in stable associations with particular sea anemones of the
species Telmatactis cricoides for up to 13 months. This shrimp species lives in
pairs as simultaneous hermaphrodites, a social structure not previously
described for any crustacean (Wirtz 1997).
2
Both mechanical and chemical stimuli are neede to release the nematocysts of
a sea anemone. The ability of a compound, from e.g. the exoskeleton of a
shrimp, to incite cnidae discharge increases with the amount of mechanical
stimulation caused by the shrimp (reviewed by Crawford 1992). Crawford
found a slight correlation between time of acclimation to the giant sea
anemone Condylactis gigantea by the shrimp Periclemenes anthophilus, and
the length of the shrimps, with shorter shrimps acclimating faster.
O’Connor et al (1977) investigated which species were associated with P.
multiplicatus in Kilkerrin Bay on the Irish west coast. They found that as
many as about 40 animal species associated with the cerianthid. The most
conspicious among these were the fan-worm Myxicola infundibulum, the
bivalvia Mysella bidentata and the sipunculid Golfingia elongata. The latter
association might have been parasitic as nematocysts from the cerianthid
were found in the gut of the sipunculid.
They proposed the suitability of the cerianthid tube as a substrate for
settlement as the main reason for all these associations, as many of the
associating species were not found anywhere else in the research area, except
on, or in, the tubes of P. multiplicatus. The research area consisted of a soft
muddy bottom without any objects to settle on. They defined the associations
as commensalistic, as they could not find any advantages for the cerianthid
from the associations (O’Connor et al 1977).
In the subtropical parts of the Eastern Atlantic and the Medierranean the
spider crab Inachus phalangium has a social structure where the females are
stationary, associated with the sea anemone A n e m o n i a sulcata ,while the
males, stationary at day, are travelling around at night visiting the stationary
females (Wirtz 1997).
Another variant of associations between sea anemones and crabs was found
by Patton (1979). Only the juveniles of the spider crab Mithrax cinctimanus
associated with sea anemones while the adults were living as non-symbionts.
The juveniles were further found to have a rather temporary association
with their host sea anemones and often moved on after a single day or a few
days.
Even so, during laboratory experiments Patton could observe that M.
cinctimanus would always move into association with one or another sea
anemone rather than hiding under stones or pieces of coral as did nonsymbiont species of spider crabs, showing that an association with a sea
anemone was strongly preferred. This preference was, however, possibly due
to the unnatural conditions in the laboratory, rather than proving a strong
association which was not confirmed by the field studies.
The normal position of the spider crab Stenorhynchus lanceolatus, associated
with the sea anemone Telmatactis cricoides at the Canary Islands, was sitting
close to the sea anemone, facing outwards, and usually in contact with the sea
anemone through placing one or two of the rear legs between the tentacles of
the sea anemone (Wirtz 1997).
3
The deep bottoms (60-250 m) in the Koster fiord area, outside the west coast of
Sweden, have been filmed with ROVs (remotely operated vehicles) during a
number of occasions over the last two years, and time after time it was
observed that shrimps of some specific species Spirontocaris liljeborgii,
Lebbeus polaris, Pandalus borealis, P. propinquus and P. montagui, aggregated
beneath B. tuediae and Pachycerianthus multiplicatus (T. Lundälv pers
comm). This suggested that associations similar to the ones between tropical
sea anemones and shrimps might exist in temperate waters as well.
However, not only shrimps were seen to associate with B. tueidae, but also
the lithodid crab Lithodes maja was quite often observed sitting beneath the
canopy of the tentacles of a B. tuediae.
In the present study, a model of a non-random distribution of the shrimp
species and the lithodid crab on the sea-bed, is made from the observations in
situ with the ROVs. A hypothesis of an symbiotic association is tested, using
the a priori knowledge from the observations, that shrimps and crabs are
aggregating beneath the two anthozoans.
The aim of the study is to investigate some of the more basic questions
concerning this newly detected association, such as which species are
involved, and to what degree, and what the possible reasons for this
symbiosis might be. Further on the associations between sea anemones and
crustaceans from temperate waters are compared with similar associations in
tropical waters as far as possible.
Materials and methods
Species
The two anthozoans, B tuediae and P. multiplicatus, are both common in the
Koster fiord where depths and bottom substrates are suitable for them.
However, B. tuediae appears to be the more common species of the two and
P. multiplicatus was only found in great numbers at certain locations, e.g.
south of Yttre Vattenholmen and at Hällsö in the northern part of the fiord.
Type of bottom substrate, but also prevailing bottom currents and the rate of
accumulation of sediments seemed to be of great importance to both
anthozoans. As a result their distribution was very patchy. At a certain point a
dense anthozoan population could be found, and yet only a few hundred
meters away only a few individuals would be found.
Only on four occasions were both species found at the same filmed location.
The reason for this rare co-occurrence was probably due to the species
preferring slightly different habitats.
B. tuediae has numerous large nematocysts on the column, on the tentacles
and even on the mesenterial filaments, and is usually described as a
voracious predator (Vader 1970a). P. multiplicatus has weaker nematocysts
and feeds mainly on plankton (pers. observation).
4
Five species of shrimps, found associated with the anthozoans, have been
identified with certainty, namely Spirontocaris liljeborgii, Lebbeus polaris,
Pandalus borealis, P. propinquus and P. montagui. Some other species, e.g. S.
spinosus, may associate with the anthozoans in question, though this species
is rarely encountered off the Swedish west coast (Hansson pers comm). If this
was the case, those species were not observed or identified during the
filmings.
For more information about the species se Appendix A.
Field studies
The field studies were conducted aboard R/V Nereus with two ROVs. The
bigger Phantom S4 has a depth range of 300 meters and a cable length of 335
meters. The smaller Phantom XTL has a depth range of 150 meters but a cable
length of only 122 meters which restricted its use to a maximum depth of
about 100 meters. Each Phantom was equipped with a depth monitor,
videocamera, head lights, compass and chronometer and was driven by
thrusters at a maximum speed of 2 knots. The Phantom XTL had a Sony EVI331 colour videocamera and the Phantom S4 had an OSPREY OE1362 colour
videocamera.
On R/V Nereus the equipment consisted of a transformer, stabilizing the
current from Nereus (and also transforming the 220 V current from R/V
Nereus to 110 V for the Phantom XTL). This transformer was connected to
the manouvering console from which the ROV was steered and the angles of
the camera and the headlights controlled. The video signal from the ROV
went to a VCR and was shown on a monitor. The cable connecting the ROV
with the manouevering console contained electric wires used to power the
camera, the head lights and the engines and sending back signals from the
video camera and the instruments.
To ensure that the filmed anthozoans were picked by chance the ROV was
steered randomly along the bottom of a chosen site as far as the cable would
allow. The area covered in this way was roughly circular. The size of the circle
depended on the depth but was usually around 1.800-2.500 square meters.
At each observed B. tuediae or P. multiplicatus the ROV was stopped and the
area immediately around the sea anemone or the cerianthid closely filmed.
When the filming was considered adequate to identify all shrimps
subjectively judged to be associated with the anthozoan, the ROV was steered
to the other side of it via a detour in order to have a complete filming of all
shrimps aggregated beneath it.
The control areas, used for comparisons, were chosen randomly from the
filming of the sea bed between the anthozoans. This was done by picking an
area, shown on the monitor when the tape was arrested after a count of five,
after leaving the last anthozoan filmed. This area was subjectively estimated
to be of the same size as the areas around the anthozoans.
Most of the filming was done with the Phantom XTL as its camera was of a
higher quality and better enabled the identification of the shrimp species.
5
With the Phantom S4 it was generally only possible to count the number of
shrimps beneath the anthozoan but not to identify the shrimps with high
enough accuracy. However, some filming was done with the Phantom S4 in
order to cover depths greater than 100 m.
Three areas in the Koster fiord were chosen as locations for the filming, one
in the south part, including Spiran and Ramskär, one in the middle,
including Yttre Vattenholmen, Berggylteskär, Kostergrund and Björns rev
(the last one a local non-official name on a deep rock) and one in the north,
including Hällsö, Tjurholmen, Kungsviksflaket and Kostersäcken (see
Appendix B). Twelve hours of filming were conducted in order to obtain
enough material for statistical purposes, though earlier filming and filming
for other purposes were used for more general observations of the
associations.
Shrimp experiments
In a 54 litre aquarium a transparent dividing glass wall was installed in the
middle of the aquarium from one end to a little more than half the length of
the aquarium. At the end of this glass wall a second short wall was placed
perpendicular to the first, in order to divide the water flow further.
Non-contaminated sea water entered on one side of the glass wall, while on
the other side water from an aquarium with either B. tuediae or P.
multiplicatus entered. The water outlet was situated close to the other facing
short side of the aquarium to create an even water flow through the
aquarium. The water in the aquarium was kept at 8-10 °C. and the room at a
constant temperature of 5 °C.
100 P. borealis and 12 S. liljeborgii were used in the experiments. Ten shrimps
were released in the facing short side and allowed to distribute by choice,
depending on if they were attracted by the chemical compounds released by
the anthozoan or not. By limiting the numbers released at the same time,
possible interactions between the shrimps were minimized. Their positions
in an area were noted after one hour as they seemed to need this time to
choose where to settle down in the aquarium. The areas were defined as: in
the release area or in one of the compartments with flows.
In this experiment only chemical compounds from the sea anemone and the
cerianthid could be used. Due to difficulties when catching the shrimps, it was
not possible to protect them against ultraviolet radiation damaging their eyes.
As a result, upon reaching the lab, they were all quite certainly more or less
blind. The eyes of deep-living crustaceans show a number of adaptations for
vision in low light including the development of a tapetum, superposition
optics and large rhabdoms (Land 1981, in Gaten et al 1990). When these
crustaceans, which naturally only encounter dim light, are raised to the
surface and exposed to sunlight, catastrophic breakdown of the photoreceptors
can occur (Nilsson & Lindstrom 1983 in Shelton et al 1985). In the case of
6
Nephrops norvegicus, exposure to direct sunlight for periods as short as 2 s
can damage the eye (Shelton et al 1985).
In a small pilot experiment shrimps were placed in an aquarium with a sea
anemone. They approached the sea anemone, presumably attracted by the
chemical compounds released by it, but not being able to see the tentacles they
bumped into them and violently pulled back, stung by the nematocysts (pers
obs). As the shrimps had no perception of the tentacles of the anthozoan, they
would not aggregate beneath it, as was observed in the field. Consequently
only the possible attraction of the shrimps to the chemical compounds,
released by the anthozoans, could be tested for.
Crab experiments
The aim of the experiment was initially to test whether the lithodid crab L.
maja was interested in finding shelter, and if so to identify the type of shelter
they would choose. Would they selectively choose B. tuediae as shelter to
gain the additional protection from the nematocysts of the sea anemone
against predators, or would any kind of shelter do?
An artifical sea anemone was placed in one end and a B. tuediae in the other
end of a 164 litre aquarium (length 135 cm, width 32 cm).
The artificial sea anemone was used to provide the lithodid crab with a
shelter very similar in shape and size to the true sea anemone but without
the added protection by the nematocysts of the sea anemone.
The base was made of concrete and buried in the coarse sand covering the
bottom of the aquarium. A thick plastic tube served as a column and two
different kinds of ropes as tentacles, one that hung down along the column
and one that floated above the column in the water. Both ropes were
unwound so as to create the impression of a multitude of tentacles.
The water flow entered the aquarium in the middle, close to the bottom and
left along the upper edges as the aquarium was allowed to overflow. A dim
light was coming from the right. All other possible environmental gradients
were considered to be negligible. The water and room temperature were the
same as for the shrimp experiments.
One L. maja at a time was put down in the middle of the aquarium facing
one of the long sides. The area of the aquarium was divided into three
sections and it was noted in which section the crab was found at every
checking time. The three sections consisted of: 1) immediately around or
beneath the fake sea anemone, 2) immediately around or beneath B. tuediae
or 3) in the middle area of the aquarium not close to any of the shelters. The
middle area was of about the same size as the two sheltered areas combined.
The position of each crab was observed after 30 min, one, two, three and
twelve hours. In the statistical testing the position after two hours was
chosen as the most representative for the purpose of the experiment, i.e. to
investigate the need for shelter of the crabs. Two hours seemed to be the time
it would take for the crabs to settle down following the initial stress of being
7
handled and moved to a new habitat, but before they would feel completely at
home in the new habitat and thereby loose their presumed need for a shelter.
When all the crabs had been tested once the positions of B. tuediae and the
fake sea anemone were switched around, as it was suspected that the light
coming from one side of the aquarium would influence the choices of the
crabs.
Only twelve crabs were used altogether as they were difficult to catch, six
females and six males. In order to identify the individual crabs they were
marked with coloured rubberbands fastened to their rear legs and their
weights were noted.
Between the tests, the crabs were kept in a big tank in the same room as the
aquarium.
Location and the interaction between location and the choices of the shrimps
Results
Field studies
Video recordings were made at three different locations in the Koster fiord
area, namely in the south, middle and north parts of the fiord. A two-way
ANOVA in an orthogonal model and with the geographic locations as a
random factor and the choices of the shrimps (all species of shrimps pooled
together) as a fixed factor, confirmed the generality and similarity of the
associations at the three locations (table 1, fig 1). The data showed
heterogenous variances according to Cochran’s test which, however,
disappeared using log-transformation.
Table 1. Two-way orthogonal ANOVA analysing the effect of geographic location (random
factor with three levels, north, middle and south part of the Koster fiord), what habitat
shrimps of five different species pooled together chose (fixed factor with three levels, B.
tuediae, P. multiplicatus and control area without any anthozoan. Mean square from the
interaction as error term), and finally the interaction of the geographic locations with the
choices of the shrimps.
Source
df
Sum of
squares
Mean
square
F-Value
P-Value
Error term
Location
2
0,580
0,290
1,257
0,3083
Residual
Choices of shrimps
2
12,154
6,077
44,556
0,0018
Loc*choices of..
Location *
Choices of shrimps
Residual
4
0,546
0,136
0,591
0,6734
Residual
18
4,153
0,231
Dependent: log number of shrimps
8
showed non significant values (p=0,31 and p=0,67 respectively, n=27). Only
the choices of the shrimps was highly significant, p=0,0018.
The post-hoc test (Student-Newman-Keuls) confirmed the significant
differences between the choices of the shrimps (table 2).
Interaction Bar Chart
Effect: location * choices of the sh…
Dependent: number of shrimps
With Standard Error error bars.
35
Cell Means of number of shrimps
30
25
south
20
middle
north
15
10
5
0
B. tuediae
P. multiplicatus
Control area
Fig 1. Graph showing how shrimps of five different species pooled together chose between the
two anthozoans and the control area at the different geographic locations.
Table 2. Student-Newman-Keuls post-hoc test showing that there was a significant difference
of how the shrimps chose between the habitats, B. tuediae, P. multiplicatus and control areas.
Vs.
Difference
Crit. diff.
Control
P. multiplicatus
1,028
0,484
S
Control
B. tuediae
1,625
0,620
S
P. multiplicatus
B. tuediae
0,597
0,484
S
S=significantly different at this level
9
A total of 96 B. tuediae and 36 P. multiplicatus were filmed. Of these only
seven and six percent respectively were found to be without associating
shrimps. Though these percentages might actually be over-estimated as some
occasional shrimps around the anthozoans, not filmed from both directions
but only passed over with the camera, easily may have been overlooked.
The different species of shrimps were analysed separately to find out which of
them that had the closest association with the anthozoans and which of the
two anthozoans that was the preferred host. The Pandalus spp, P. borealis, P,
propinquus and P. montagui were pooled together in the statistics, as they
could not be separated with certainty on the video tape.
A two-way orthogonal Anova with both host species and shrimp species as
fixed factors, showed that there was an interaction between host species and
shrimp species, p=0,0042, n=54 (table 3). Thus the significant values for host
species and shrimp species were of no interest.
Table 3. Two-way orthogonal ANOVA analysing the effect of host species (fixed factor with two
levels, B. tuediae and P. multiplicatus) and shrimp species (fixed factor with three levels S.
liljeborgii, L. polaris and Pandalus spp), and the interaction between host species and shrimp
species.
Source
Host species
1
Sum of
squares
4,807
Shrimp species
2
58,807
Host species*Shrimp species
2
48
Residual
df
Mean
square
4,807
F-Value
P-Value
12,347
0,0010
29,403
75,525
0,0001
4,786
2,393
6,147
0,0042
18,687
0,389
Dependent: Square number of shrimps
The significant value for the interaction was probably due to that all shrimp
species preferred B. tuediae before P. multiplicatus but to a different degree,
e.g. S. liljeborgii associated ca. three times more with B. tuediae than with P.
multiplicatus, while L. polaris associated as much as about 50 times more
with B. tuediae than with the cerianthid (fig 2). The data showed
heterogenous variances according to Cochran’s test and were thus square-root
transformed to make the variance homogenous.
As there was no difference between the geographic locations, they were
pooled together, and instead divided into two kinds of habitats, one with a
dense population of the anthozoan in question and one with a sparse
population of them. An orthogonal two-way ANOVA with both the choices
of the shrimps and the kind of habitat as fixed factors, was conducted to see if
10
Interaction Bar Chart
Effect: Host species * Shrimp species
Dependent: Number
With Standard Error error bars.
Cell Means of Number
10
8
B. tuediae
6
P. multiplicatus
4
2
0
S. liljeborgii
L. polaris
Pandalus spp.
Fig 2. Graph showing mean number ot the shrimp species, S. liljeborgii, L. polaris and Pandalus
spp observed associating with the two anthozoans, B. tuediae, n=52, and P. multiplicatus, n=29.
the means of the number of shrimps associating to the anthozoan would
differ between these two kinds of habitats. The result gave a non significant
result for any difference between them, p=0,89, n=75 (table 4). The data
showed heterogenous variances according to Cochran’s test and were thus
log-transformed.
The lithodid crab L. maja was often seen associating with B. tuediae. In the
video recordings where both species could be observed in the same area 72
percent of the crabs were seen associating to a B. tuediae .
11
Table 4. Two-way orthogonal ANOVA analysing the effect of two kinds of habitats (fixed
factor with two levels, sparse or dense populations of anthozoans) and the choices of the
shrimps (fixed factor with three levels, B. tuediae, P. multiplicatus and control area without
any anthozoan) showed a non-significant result, p=0,89, n=75.
Source
df
Sum of squares
Mean square
F-Value
P-Value
Choice of shrimps
2
25,759
12,879
75,044
0,0001
Type of biotope
1
0,003
0,003
0,019
0,8911
Choice of shrimps*
type of biotope
Residual
2
0,395
0,198
1,152
0,3221
69
11,842
0,172
Dependent: log number of shrimps
Shrimp experiments
The fundamental presumption was that if the shrimps were not attracted to
the flows containing chemical compounds from B. tuediae or P. multiplicatus
they were expected to move randomly and distribute evenly among the three
compartments of the aquarium made up by the release area and the two flow
areas. These three areas were of about equal sizes.
First the shrimps were tested to see if the light coming from one direction
affected their choices. A blind test without any anthozoan was conducted. The
light was first coming from the same direction as the flows and then from the
same direction as the area of release by turning the aquarium 180 degrees. A
Fisher exact test gave p=1,00, n=20. Clearly the shrimps were not affected by
the light, indeed an expected result as they probably were more or less blind as
a consequence of being captured during day-time.
The next experiment tested if blind shrimps were willing to move when
attracted by the chemical compounds released by B. tuediae or P.
multiplicatus. A chi-square test gave the result p=0,013, n=92, which indicated
that more shrimps stayed in the release area (result 44 of 92, expected 30,7 of
92) than expected by chance alone.
Because of this result the remaining statistical tests only included the shrimps
that actually had moved into one of the two compartments with flows, as the
shrimps that did remain at the release area were considered less mobile than
expected.
The two shrimp species, P. borealis and S. liljeborgii were tested separately
against the two anthozoans to see if these would attract them in a varying
degree. The shrimps chose between pure sea water and sea water deriving
from the aquarium with either of the anthozoans. S. liljeborgii was attracted
12
by water from B. tuediae (p=0,045, chi-square test, n=4) while all the other
combinations of shrimp and anthozoan did not deviate significantly from a
random choice. However, the significant p-value for the test with S.
liljeborgii was a bit suspect because of the very small material, though in
accordance with the results from the field studies. P. borealis tested against
both the anthozoans and S. liljeborgi tested against P. multiplicatus showed
non-significant results.
The results from the sea anemone and the cerianthid were then pooled
together to see if there was a difference between the two shrimp species in
how often they preferred the water flow from any of the anthozoans. (fig 3). S.
liljeborgii chose the compartments with flows from one of the anthozoans
more often than expected (p=0,025, chi-square test, n=5), while this was not
the case with P. borealis.
The two species of shrimps were then pooled together but with the
anthozoans separated to see which anthozoan that was preferred in general by
the shrimps. The results though were highly non-significant, p=0,70, n=22,
for B. tuediae and p=0,40, n=13 for P. multiplicatus.
100%
90%
80%
70%
60%
Pure water
P. multiplicatus
50%
B. tuediae
40%
30%
20%
10%
0%
P.borealis
S. liljeborgii
Fig 3. Graph showing how P. borealis (n=30) and S. liljeborgii (n=5) chose between the flow
from one of the anthozoan or pure sea water flow. The graph is in percent as the number of
shrimps of the two species were unequal.
13
The shrimps were also tested so that they could choose directly between the
two anthozoans without any possibility to choose pure sea water, but this did
not give any significant difference, Fisher exact test p=1,00, n=13.
In another test where the choices of P. borealis, n=20, and S. liljeborgi, n=7,
were directly compared to each other instead of compared to expected choices,
Fischer’s exact test gave the result p=0,185, i.e. there was no significant
difference between the choices of the shrimp species. Even though the result
was non significant the graph in fig 4 shows that there might be a weak trend
that P. borealis more often would chose the flow from P. multiplicatus, while
S. liljeborgii more often would chose the flow from B. tuediae. This result
disagreed, though, with the results from the field studies, which gave the
result that all shrimp species preferred the sea anemone before the cerianthid.
14
12
10
8
B. tuediae
P. multiplicatus
6
4
2
0
P. borealis
S. liljeborgii
Fig 4. Graph showing the number of the shrimps P. borealis (n=20) and S. liljeborgii (n=7)
choosing the water flows from either B. tuediae or P. multiplicatus.
Crab experiments
The light was coming from one direction in the laboratory room, which
might have influenced the choices of the crabs in the two set-ups of the
experiment, one where B. tuediae was situated closest to the light source and
one where the sea anemone was situated furthest away from it. This was
tested for in a likelihood ratio chi-square test, which compared the results
directly between the two set-ups. All the time periods, 30 min, one, two, three
and twelve hours were compared between the two tests. Only three hours
14
gave a significant result p=0,05, while twelve hours showed a trend with
p=0,09.
Further on the different time periods were tested directly against each other
in a likelihood ratio chi-square test p=0,087, n=12. As there was no significant
difference between the the two set-ups in the time-period two hours, neither
any significant difference between the time periods, the results of the crabs
were hence chosen randomly, six crabs from each of the two set-ups and from
the time period two hours, for the remaining statistical tests.Two hours was
chosen as the most representative of the choices of the crabs. Before two
hours they were perceived as too unsettled in their new environment and
after two hours they seemed to feel so secure that they lost their interest for
finding a shelter.
If the crabs distributed at random in the aquarium or not was tested for. A
non-random distribution would suggest a habitat choice of the crabs. The
results from twelve replicate trials showed that the crabs indeed chose to sit
close to a shelter, i.e. either a true or a fake sea anemone, p=0,02, n =12.
However, when females and males were tested separately, it became evident
that this significance solely was caused by the choices of the females, p=0,01
n=6, compared to the choices of the males p=0,41, n=6 (table 5).
Table 5. Shows how the crabs total, and divided into males and females chose between shelter
(either a fake or a true sea anemone) and no shelter, and compared to a random even
distribution. Total crabs n=12, males n=6, females n=6. P-values given from chi-square tests.
Total crabs
Observed
Shelter
No shelter
P-Value
Females
Expected
Observed
Males
Expected
Observed
Expected
10
6
6
3
4
3
2
6
0
3
2
3
0,02
0,01
0,41
The last test, was to find out if the crabs actively would choose B. tuediae for a
shelter, or if any shelter would do. A chi-square test compared how the crabs
chose between the B. tuediae and the fake sea anemone and compared it to
the expected numbers if the crabs had chosen randomly between the two
kinds of shelters. For all crabs p=0,003, n=10, for female crabs p=0,004, n=6, and
for male crabs p=0,197, n=4. Not surprisingly, considering the earlier result,
again the significance for all crabs turned out to be entirely due to the choices
of the females (table 6)
15
Table 6. Shows how the crabs total, and divided into females and males chose between the true
sea anemone and the fake sea anemone, and compared to a random even distribution. Only the
crabs that chose shelter, either B. tuediae or the fake sea anemone are shown here. Total n=10,
females n=6, males n=4. Only six crabs were expected to seek shelter of any type. P-values
given from chi-square tests.
Total crabs
Observed
Females
Expected
Observed
Males
Expected
Observed
Expected
B. tuediae
8
3
5
1,5
3
1,5
Fake
2
3
1
1,5
1
1,5
P-Value
0,003
0,004
0,197
Discussion
Field studies
B. tuediae was clearly the most popular host of the two anthozoans for all the
observed shrimp species. At the four occasions where both anthozoans were
found together at a filmed location, more shrimps were always gathered
around B. tuediae than around P. multiplicatus.
There may be several reasons for this preference; B. tuediae was the most
common anthozoan of the two, it has the strongest nematocysts and will thus
provide a more secure shelter than P. multiplicatus. It is also assumed to be
extremely long-lived, in accordance with many other actiniarians (Vader
1970a). In New Zealand there is an aquarium record of a small sea anemone
more than 300 years old! It is likely that many specimens of the so called
“gigantic” tropical sea anemones are at least a century old (Fautin 1992). No
information of the longevity of P. multiplicatus was available.
All of the above factors fit very well with the demands Roughgarden (1975)
discusses as necessary conditions for a symbiotic association to evolve and for
the dependency upon the host by the symbionts. According to him the initial
formation of an association occurs when the fitness of a symbiont strategy
exceeds that of a free-living or solitary strategy.
For a symbiosis to evolve, three factors must occur:
1. the host should be easy to find
2. the host should survive well with the symbiont
3. the host should provide substantial benefit to the guest.
Furthermore, host dependency increases with the importance and size of the
benefits the host provides and the longevity of the host (Roughgarden 1975).
The second point in the list above is valid only for mutualism and
commensalism, and not parasitism, but probably Roughgarden only included
these two forms of symbiosis in his definition.
16
It would have been interesting to investigate if the shrimp species in this
survey would associate with other anthozoans more common than B. tuediae
and P. multiplicatus in other geographical areas. In the deeper parts of the
Koster fiord B. tuediae and P. multiplicatus are the two most common
anthozoans. The few sea anemones of other species observed during the
filmings were never seen with any shrimps associating to them.
The few B. tuediae observed without any shrimps associating to them may
have been specimen recently opened up after a period of having had the
tentacles withdrawn into the column. Occasionally, individuals of the sea
anemone with completely withdrawn tentacles were observed when filming.
These withdrawn sea anemones would not provide the shrimps with any
significant shelter, and a guess is that the shrimps rather quickly would leave
them, trying to find an active sea anemone.
On some occasions, when shrimps were missing underneath the sea
anemone, it could simply have been caused by a total absence of any shrimps
locally in the area. No information on how long distances the shrimps would
travel on the sea bed was available, and so there was no possibility to draw
any conclusions of how long a distance they would search for one of the
anthozoans in order to find shelter. Or if they would search at all, unless they
were able to pick up chemical compounds from another anthozoan.
P. multiplicatus is not able to withdraw its tentacles but instead it withdraws
the column into the tube when disturbed so that the tentacles rest directly on
the bottom substrate and leave very little space for any shrimps to hide
beneath them. In extreme cases the cerianthid is able to withdraw completely
down into the sea bed (Lundälv pers. comm.)
Neither B. tuediae nor P. multiplicatus were perceived to gain any advantages
from their associations with the shrimps or crabs during this study, so it is
suggested that this association is commensalistic. Whether the shrimps
gained any other advantages than protection from predators was not possible
to deduce after this short study. Though quite often shrimps could be seen
picking in the sediment beneath B. tuediae with their pereiopods, presumably
finding food particles left from the feeding of the sea anemone or regurgiated
by it, thus easy access to food may be another advantage for the shrimps.
Crawford (1992) found that many tropical shrimps have an obligate
association with their host sea anemones. This is not true of the shrimp
species associating to B. tuediae or P. multiplicatus in Swedish waters. They
are all found living as non-symbionts as well. Pandalus borealis and P.
propinquus are very common on all types of deep bottoms. The other species,
S. liljeborgii and L. polaris, were rarely observed during filming except
beneath B. tuediae or P. multiplicatus. This gave a strong indication of the
strength of the association for these species. But as long as even a very few
individuals of these species actually were observed in other places than
around any of the anthozoans, the association can not be defined as obligate,
but facultative. However, it might be speculated that the few specimen of S.
liljeborgii and L. polaris observed not associated with one of the anthozoans,
were individuals searching for a new host.
17
Most of the shrimp species associating with the anthozoans maintained
rather distinct positions in relation to their host. S. liljeborgii was observed
closest to the column, and in many cases even sitting on the column. In a few
cases the shrimps were even filmed sitting among the tentacles. Pandalus spp,
on the other hand, were typically found sitting in a ring around the sea
anemone outside S. liljeborgii. Only L. polaris, seemed to be flexible in their
choice of distance from the column.
Wirtz (1997) investigated for how long individuals of the shrimp L. grabhami
would stay with their particular host sea anemone. He found that the same
individual shrimp could associate with a specific sea anemone for up to 13
months. If the same individual shrimp of the species found in the Koster
fiord, would have been found around their specific anthozoan a day later, a
week later or a month later from the first video recording, would have been
interesting to find out, but for technical reasons this was not feasible.
Though not significant, slightly more shrimps seemed to aggregate around
the anthozoans in habitats with sparse populations of these. A rather logical
observation, with few anthozoans the shrimps would have to aggregate in
larger numbers around the few available to them.
Often when filming a B. tuediae there was an abundance of krill,
Meganyctiphanes norvergica, swimming around the sea anemone, perhaps
attracted by the headlight from the ROV. They were probably temporarily
blinded by the light and swam straight into the tentacles of the sea anemone,
where they immediately were stunned by the nematocysts. The tentacles
reacted strongly to the caught krill by quickly curling the tips to ensure the
catch.
Sometimes S. liljeborgii were observed to become entangled in the tentacles
of the sea anemone when they backed away from the ROV or were disturbed
by the blinded krill. When, upon contact, the shrimps were stung by the
nematocysts of B. tuediae they would violently jerk backwards and then
directly again assume their normal position, indicating that they were not
paralyzed by the discharge of the nematocysts. The sea anemone would
typically react to the contact with the shrimp by only a slight contraction of its
tentacles, in clear contrast to when the krill were captured by it.
The most important predators on shrimps are the cod (Gadus morhua), the
haddock (Melanogrammus aeglifinus) and the whiting (Merlangius
merlangus). P. borealis is especially predated upon by the hagfish (Myxine
glutinosa), while the eel-pout (Zoarces viviparus), guillemots and seals eat P.
montagui in large numbers in areas where these are found in shallow waters
(Smaldon 1979). The populations of cod, haddock and whiting in Swedish
waters have decreased because of intensive industrial fishing. This could
possibly lead to that the association between anthozoans and shrimps would
loose its importance for the shrimps. But smaller fishes like the poor-cod
Trisopterus minutus and the Norway pout Boreogadus esmarki have instead
increased in numbers. They are also predators on shrimps, so the need for
protection against predators has not diminished for the shrimps, only the
species of the predators have changed.
18
The results from the field studies suggests that both S. liljeborgii and the
Pandalus spp. pick B. tuediae as a first hand host, presumably because of their
strong nematocysts, though the Pandalus spp. keep a somewhat longer
distance to the tentacles than S. liljeborgii.
L. maja was often found beneath the canopy of the tentacles of a B. tuediae,
but was never observed beneath a P. multiplicatus. If this was because L.
maja and the cerianthid were allopatric, or if the crab did not approve of P.
multiplicatus as a shelter could not be concluded, though the latter seems to
be more likely. Another possibility might be that P. multiplicatus is so
sensitive to touch that it immediately contracts into its’ tube when L. maja try
to position itself beneath it.
Predators on L. maja may be octopuses which are well-known for their
enthusiasm of crustaceans as preys. They are also sensitive to the nematocysts
of anthozoans, as shown by the protection hermit crabs are provided by their
associations with sea anemones they are carrying on their shells (McClean
1983, Ross 1971). A hypothesis is that the nematocysts of P. multiplicatus are
too weak to discourage the octopuses, which would then explain why L. maja
only associates with B. tuediae with its stronger nematocysts.
As a rule the crab was always backed up towards the column facing outwards.
Their rear end was pressed against the column while their rear legs were
circling the column. The same behaviour was shown by the spider crab S.
lanceolatus (Wirtz 1997). A maximum of three lithodid crabs have been
observed sitting around the same sea anemone twice, though a single crab
was the most common number.
The lithodid crab L. maja did not seem to be percieved as a threat to the
shrimps sharing the same sea anemone with them, as they were often seen
sitting underneath the crab, on top of the carapax, or walking by very close to
the claws of the crab.
While the shrimps were never seen chased away from the sea anemone
during a filming sequence, L. maja would sometimes, after usually only a
short time of filming, leave its typical position at the column of the sea
anemone and walk away some distance. If it found the light from the ROV
disturbing or if it perceived B. tuediae as an ineffective shelter against the
ROV was difficult to decide.
Patton (1979) could observe how juveniles of M. cinctimanus had rather
short-lived associations with their host sea anemones. How long a L. maja
would stay beneath a particular sea anemone was not possible to find out as
the same location was never filmed more than once. Even if it had been
possible, the question remains how much the ROV actually disturbed the
crabs.
In the field studies, some L. maja were seen sitting close to stones, on cliff
walls or close to the sponge Geodia baretti but only if a B. tuediae was not seen
nearby. These observations strongly indicate that the lithodid crab indeed
preferred B. tuediae as a shelter against predators and only would choose
stones or sponges as a substitute when no B. tuediae was found in the
immediate area.
19
Shrimp experiments
The design of the experiments on shrimp behaviour was essentially the same
as that used by Guo et al (1996). A comparison of the results was thus possible.
The experiments were conducted with P. borealis and S. liljeborgii in order to
find out if they would locate their host sea anemones in a way similar to the
tropical shrimp species, i.e. through detection of the chemical compounds
released from the anthozoans.
Unfortunately, as the shrimps were completely blind and very probably not
behaving normally, the results from the experiments are a bit dubious. Close
to half of the shrimps did not move from the release place at all, significantly
more than expected if they had distributed evenly in the aquarium without
being attracted to the two anthozoans. It might be argued, that as they were
blind, and thus probably unwilling to move at all, the attraction of the
chemical compounds from the anthozoans must have been strong indeed to
attract even one third of the shrimps.
Crawford (1992) found a certain correlation between the size of the shrimp
and the acclimation time to its host sea anemone. This correlation might be
one explanation why the smaller and shorter S. liljeborgii were found to have
a closer association to B. tuediae, they simply would not cause such a big
discharge of the nematocysts as the Pandalus spp. A fact that might have
facilitated and strengthened the association into a closer one compared to the
association between B. tuediae and the Pandalus spp.
The results from the aquarium experiments did not show that one anthozoan
was favoured before the other as the field studies clearly did, where B. tuediae
was the preferred host species. The unnatural conditions during the
experiments and the low number of shrimps might explain why it was not
possible to detect any difference in the experiments.
Crab experiments
Only the juveniles of the spider crabs Mithrax cinctimanus and Hyas araneus
have been found to associate with sea anemones while the adults of these
species are non-symbionts (Patton 1979). Unfortunately it was not possible to
find any information about at what size L. maja starts to reproduce and
thereby can be considered as an adult. This made it impossible to test for any
differences between juveniles and adults concerning their liability to associate
with a B. tuediae. During the laboratory experiments the two biggest female
crabs carried eggs on two occasions. Both were found beneath the B. tuediae
when tested, which may indicate that adult L. maja in fact associate with B.
tuediae.
I. phalangium eats food regurgiated by the sea anemone and mucus from the
surface of the sea anemone and even crops the tips of the tentacles of the sea
anemone (Wirtz & Diesel 1983 in Wirtz 1997). Whether L. maja does the
same was never observed during the filming or the laboratory experiments.
20
The females of I. phalangium would stay with their particular host sea
anemones while the males would be moving around at night searching for
the females (Wirtz 1997). If L. maja employed the same social structure could
not be investigated as all filming was conducted during daytime.
Furthermore, it was not possible to identify the gender of the crabs filmed that
were found sitting close to B. tuediae or moving around on the bottom.
But the crab experiments, though a small material, gave a clear indication
that the female crabs were much more inclined to seek protection beneath a
shelter, and then especielly beneath a B. tuediae, than the males. Possibly, this
suggests that L. maja may have a similar social structure as I. phalangium has.
Acknowledgments
First of all many thanks to my supervisor Tomas Lundälv for all the help and
time you put into this work.
You are a wonderful ball wall to throw ideas and questions at, Kerstin
Johannesson. Thank you for being around, always so helpful and interested
and taking your time with a student!
Thanks and thanks again for guiding me through the maze of statistics, Per
Nilsson!
Thank you for all the nice days spent aboard Nereus, Ingmar and Harald.
Your never ending patience and good moods made the sometimes long days
seem short.
A thousand thanks Allison for correcting my swenglish mistakes.
A very special thank you Jan Nilsson. You were the first to open my eyes to
the fascinating world of ecology, even though you are “only a birdman” and
not a marine biologist (just joking Jocke)!
21
References
Crawford, J. A. 1992. Acclimation of the shrimp, Periclimenes anthophilus, to
the giant sea anemone, Condylactis gigantea. Bulletin of marine science, 50
(2). pp 331-341.
Dahl, E. 1960. In Djurens Värld, Ryggradslösa djur. Ed. Hanström, B.
Förlagshuset Norden AB, Malmö. Pp 315-331.
Fautin, D. G. 1992. Field guide to anemone fishes and their host sea
anemones. Western Australian Museum, Perth, Australia.
Fautin, D. G. Guo, C-C. Hwang, J-S. 1995 Costs and benefits of the symbiosis
between the anemoneshrimp Periclimenes brevicarpalis and its host
Entacmaea quadricolor. Marine Ecology Progress Series, vol 129, pp 77-84.
Gaten, E. Shelton, P. M.J. Chapman, C. J. Shanks, A. M. 1990. Depth related
variation in the structure and functioning of the compound eye of the
Norway lobster Nephrops norvegicus. Journal of marine biology, vol 70, pp
343-355.
Guo, C-C. Hwang, J-S. Fautin, D. G. 1996. Host selection by shrimps symbiotic
with sea anemones: a field survey and experimental laboratory analysis.
Journal of Experimental Marine Biology and Ecology, vol 202, pp 165-176.
Hansson, H. G. 1998. Sydskandinaviska marina flercelliga evertebrater
(utgåva 2). Länsstyrelsen Västra Götaland.
Knowlton, N. 1980. Sexual selection and dimorphism in two demes of a
symbiotic, pair-bonding snapping shrimp. Evolution, 34(1). pp 161-173.
Lagerberg, T. 1908 Sveriges decapoder. Wald Zachrissons boktryckeri AB,
Göteborg, page 62.
Manuel, R.L. 1981. British Anthozoa, Synopsis of the British Fauna no 18. Ed
D.M. Kermack, R.S.K. Barnes, Academic Press, London, pp 67, 104.
McClean, R. 1983. Gastropod shells a dynamic resource that helps shape bentic
community structure. Journal of experimental marine biology and ecology 69
(2), pp 151-174.
O’Connor, B. et al. 1977. Pachycerianthus multiplicatus – biotope or
biocoensis? In Biology of Benthic Organisms. Ed B.F. Keegan & P.O. Ceidigh,
P.J.S. Boaden. Pergamon Press Ltd, Oxford. pp 472-482.
Patton, W. K. 1979. On the assocation of the spider crab, Mithrax
(Mithraculus) cinctimanus with Jamaican sea anemones. Crustaceana, suppl
5. pp 55-61.
22
Ross, D. M. 1971. Protection of hermit crabs (Dardanus spp) from octopus by
commensal sea anemones (Calliactis spp). Nature vol 230, april 9, pp 401-402.
Roughgarden, J. 1975. Evolution of marine symbiosis – a simple cost-benefit
model. Ecology vol 56, pp 1201-1208.
Shelton, P. M. J. Gaten, E. Chapman, C.J. 1985. Light and retinal damage in
Nephrops norvegicus. Proc. R. Soc. Lond. B 226, pp 217-236.
Shick, J. M. 1991. A functional biology of sea anemones. Chapman & Hall,
London. pp 22-24.
Smaldon, G. 1979. British Coastal Shrimps and Prawns, Academic press Inc,
London.
Smith, W. L. 1977. Beneficial behaviour of a symbiotic shrimp to its host
anemone, Bulletin of marine science, vol 27, no 2, pp 343-346.
Vader, W. 1970a Amhipods associated with the sea anemone, Bolocera
tuediae, in western Norway. Sarsia 43, pp 87-98.
Vader, WW. 1970b. Antheacheres duebeni, a copepod prasitic in the sea
anemone Bolocera tuediae. Sarsia 43, pp 99-106.
Wirtz, P. 1997. Crustacean symbionts of the sea anemone Telmatactis cricoides
at Madeira and the Canary Islands. Journal Zoology London, 242, pp 799-811.
23
24
Appendix A
Map of the Koster fiord area. Field studies were conducted in three areas.
The north part of the fiord including Hällsö, Tjurholmen, Kungsviksflaket
and Kostersäcken.
The middle part including Yttre Vattenholmen, Berggylteskär, Kostergrund
and Björns rev.
The south part including Spiran and Ramskär.
25
Appendix B
Descriptions of the species in the study
Bolocera tuediae
This sea anemone is found from 20 m down to more than 2000 m and is
common on both sides of the northern Atlantic as far north as to the arctic
cirle (Manuel 1981).
The colour of the column is usually light red, yellowish orange or sometimes
white with the tentacles usually of the same colour as the column but slightly
darker and often with faint stripes (Vader 1970). Occasionally specimens with
dark purple tentacles were observed during the video recordings. The
tentacles are deciduous and are readily shed on occasion, the autotomized
tentacles can remain alive for weeks although they are not known to be able
to regenerate into new sea anemones (Manuel 1981). The reason for this
shedding is unknown but defense against predators have been suggested. The
sea anemone has numerous large nematocysts on the column, the tentacles
and even on the mesenterial filaments, and it is the only sea anemone in
Swedish waters that can be painful to humans (Vader 1970).
The maximum size is about 25 cm. It is probably an extremely long-lived
species where the size of mature specimens are dependent mainly on the
availability of food (Vader 1970).
The sea anemone is dioecious, the reddish eggs, with a diameter of about 1.1
mm, are among the largest anthozoan eggs known. The pelagic larval
develoment takes about one month. In deep water reproduction occurs
throughout the year but in more shallow water the reproduction is probably
restricted to spring (Vader 1970).
It is usually found on soft sediments mixed with pebbles, shells, stones and
cliffs, where it is able to attach itself to the hard substrates, though unattached
specimens are observed as well (Vader 1970).
B. tuediae is usually described as a voracious predator and crustaceans like
Pontophilus, Munida, Pandalus, Mysidacea have been found in the
gastrovascular cavities. Once the capture of a small fish, about 10 cm in
length, was observed during the video recordings. In the laboratory B. tuediae
readily accepted fish meat and mussel meat (Vader 1970).
Pachycerianthus multiplicatus
This cerianthid species was not found until 1912 by O. Carlgren in the
Trondheim fiord (Dahl 1960).
It is found from 15 m and down to more than 200 m depth at the west coasts
of Scandinavia and around Scotland and Ireland (Manuel 1981).
The cerianthid lives on soft muddy sediment bottoms where it is able to dig
into the substrate with its tube to anchor itself. The larvae settle after a couple
of days from a short pelagic life (Dahl 1960).
The column is often salmon pink and the long whitish tentacles around the
edge of the disc are distinctly striped with brown stripes even though
completely white specimens have been recorded. The short tentacles around
26
the mouth are of a single greyish brown colour (Hansson 1998). The
maximum size is about 30 cm with the tube an additional 130 cm of which
only about ten cm projects above the sediment (O’Connor 1977).
The tube is built by the animal itself. The outer layer (3-7 mm thick) consists
of sedimentary debris and is ramified by filamentous algae, the inner layer (24 mm thick) is composed of mucous strings and nematocysts derived from
the cerianthid itself (O’Connor 1977). The colour of the tube is the same as the
sediment, i.e. gray or black.
A very interesting atelia was frequently observed in P. multiplicatus when
filming them. Very often the cerianthid would slowly turn its tentacle crown
towards the direction of the headlights of the ROV, quite similar to a flower
turning towards the direction of the sun. Obviously it was able to sense the
light. An intriguing fact as the cerianthids usually are found at depths where,
at most, only the tiniest amount of sunlight is able to penetrate.
Sea anemones are known to respond to light even though no photoreceptor
structures or pigments ever have been identified in them (Shick 1991). Local
illumination of the column of Metridium senile caused contraction of the
parietal muscles in the illuminated (reviewed by Shick 1991). Further
research done by North and Pantin (1958 in Shick 1991) and Marks (1976 in
Shick 1991) could show that no chemical synaptic transmissions or electrical
activity could be recorded during the contractions. Thus it seemed likely that
the myoepithelial cells themselves were photosensitive (Shick 1991). As sea
anemones and cerianthids are closely related, the same photosensitive system
in cerianthids might be assumed.
The question remains, though, for what purpose P. multiplicatus uses its
light sensitivity! The sea anemone B. tuediae, also a deep bottom species like
P. multiplicatus, has not showed any sign of reaction towards the light when
illuminated by the headlights.
Spirontocaris liljeborgii
This shrimp is usually found between 20 and 230 m but has been found as
deep as 1200 m. Distribution range is Iceland, Greenland, British Isles,
Scandinavia and Massachusetts, North America. Maximum size is 7.4 cm, but
it is usually between 4 and 6 cm with the females usually larger than the
males. They are found on both hard and soft bottoms. Breeding occurs during
winter. The colour is bright red or sometimes dark red. The rostrum has a
convex upper margin. This species eats large quantities of foraminiferans
(Smaldon 1979).
Pandalus borealis
This is the common commercial shrimp (Nordhavsräkan). It is found
between 20 and 900 m depth, frequently between 80 and 500 m. Maximum
length is 16 cm, but it is usually about 10 cm. The colour is pale red, and it
lives mainly on muddy substrates. P. borealis is a protandric hermaphrodite
and breeds during winter. Known from the north Atlantic and the Pacific
Ocean as far south as 41 degrees North (Smaldon 1979).
27
Pandalus montagui
This shrimp is also called the Caramel shrimp as it is semi-translucent with
red stripes. It is found between 5 and 230 m depth, though seldom deeper
than 100 m. The maximum length is 16 cm, but it is usually less than 10 cm. It
lives on gravelly, sandy and muddy substrates. It is a protandrous
hermaphrodite, males usually developing female characters after 13 to 16
months. They breed during winter. Known from the north Atlantic and
north-east Pacific Ocean as far south as 39 degrees North. It feeds largely on
the polychaete worm Sabellaria spinulosa, but small crustaceans,
foraminiferans, hydroids and other polychaetes have been found in the
stomach as well (Smaldon 1979).
Pandalus propinquus
It lives from 40 m down to 2000 m depth on hard and soft bottoms. It is red
and has a maximum size of 15 cm, though it is usually less than 10 cm. The
colour is pale red. This species is dioecious in contrast to the other two
Pandalus species and it breeds during winter. Its rostrum is strongly curved
upwards while the rostrum of P. borealis is straight and that of P. montagui is
slightly curved upwards. Its range is north-east and north-west Atlantic down
to 39 degrees North (Smaldon 1979).
Lebbeus polaris
This shrimp is found from 0 m down to 930 m, but usually between 30 and
300 m. The maximum length is 9 cm, though it is usually between 6 and 7 cm.
It is transparent with red dots and conspiciously red and white striped legs.
Occurs on both hard and muddy substrates. Its range is circumpolar. The
breeding season differs in different parts of its geographical range (Smaldon
1979).
Lithodes maja
This crab belongs to the family Lithodidae with only this single genus in
Swedish waters. DNA-sequencing shows that lithodid crabs are closer related
to hermit crabs than to true crabs (pers. comm. H. G. Hansson).
L. maja is found from 10 m. down to 500 m, though usually between 30 and
200 m, on both hard and soft bottoms. Their maximum size is about 15 cm
across the carapax. The colour is reddish or reddish brown. Favourite food is
brittle stars. The last pair of legs are strongly reduced and difficult to see
(Hansson 1998). Carapax and legs are covered with spines, with the juveniles
spinier than the adult specimens (Lagerberg 1908). The abdomen is small and
slightly asymmetrical, more so in females, reflecting the pagurid ancestry
(Hansson 1998). Their distribution range is the north-east Atlantic coast and
the coast of east North America.
28